Stereo Matching
Stereo matching aims to reconstruct 3D depth from two or more images by identifying corresponding pixels across different viewpoints. Current research heavily utilizes deep learning, focusing on improving accuracy and efficiency through novel architectures like Transformers and convolutional neural networks, often incorporating iterative refinement and uncertainty estimation techniques. This field is crucial for applications such as autonomous driving, robotics, and 3D modeling, with ongoing efforts to enhance generalization across diverse datasets and improve real-time performance on resource-constrained devices.
Papers
February 28, 2024
February 14, 2024
January 21, 2024
January 17, 2024
January 9, 2024
January 3, 2024
December 26, 2023
December 22, 2023
December 14, 2023
December 1, 2023
November 4, 2023
October 3, 2023
September 8, 2023
September 4, 2023
August 30, 2023
August 6, 2023
August 1, 2023
July 31, 2023
July 26, 2023